How to Build an Investor List for a Startup Fundraise
Build a focused investor list by stage, sector, check size, geography, thesis and contact quality — and learn how short the list should actually be.
A long list is not an asset
The standard advice is to build a list of two hundred investors and work down it. It sounds like diligence. In practice it is the single most expensive mistake in a first raise, because it converts a targeting problem into a volume problem and then measures the wrong thing.
We keep a dataset of 1,711 venture firms, each with the investment thesis they publish on their own website. When we counted how many sectors those firms name, the answer was blunt: 1,417 of them — 83% — name three sectors or fewer. Only 113 claim six or more.
So most firms are narrow, and they say so publicly. A list of two hundred names assembled without reading those theses will be mostly firms that were never going to invest in your category. When they do not reply, that reads as rejection. It was a category error.
Six things that actually qualify an investor
A name belongs on your list only once you can answer these. If you cannot, you are guessing.
- Stage. Pre-seed, seed, Series A, growth, or opportunistic. A seed fund does not write pre-seed cheques as a favour.
- Sector. Does their stated thesis include what you are actually building, in their words rather than yours?
- Cheque size and role. Do they lead, co-lead, follow, or write angel-sized cheques? A follower cannot set your terms.
- Geography. Some funds are strict about jurisdiction, some are not. This is usually written down.
- Portfolio context. Adjacent investments are a signal. A direct competitor is a conflict and usually a hard no.
- Contact quality. Is there a named partner, or only a general inbox?
Access is rarely the real problem
Founders often assume the hard part is reaching anyone at all. The data does not support that. Of those 1,711 firms, 1,710 publish at least one named contact address, and 1,439 — 84% — name a partner or principal specifically.
A warm introduction is still better than a cold email. It is not a prerequisite. The binding constraint is almost never "I cannot reach a partner". It is "I cannot tell which of these partners would care".
How long should the list be?
Shorter than you think, and it should get shorter as you qualify. Thirty firms you can each justify in one sentence will out-perform two hundred you cannot, because the justification is the same thing you need for the first line of the email.
If you are building in a narrow category this feels alarming. Forty-two firms in our dataset name gaming, against 534 that name AI. If you are building a game, your addressable universe is roughly twelve times smaller than an AI founder's. That is not worse — it is knowable. You can read all forty-two theses in an afternoon. Nobody can meaningfully read 534.
Where software helps, and where it does not
Matching is a reading problem at heart: comparing what you are building against what each firm says it wants. That is worth automating, because it is repetitive and it scales badly by hand.
Deciding who to contact, and what to say to them, is not. That judgement is the part investors are actually responding to.
Roundly publishes your round as a page and matches it against investor theses, so discovery runs on stated fit rather than on list length. You approve every introduction. You decide who you meet.
Bottom line
Build the shortest list you can defend. Read the thesis before the name goes on it. Volume only helps after targeting is right — before that, it just scales the damage faster.
Run your fundraise on Roundly
Publish your round once. Investors who have defined a thesis discover it, and you approve every introduction — you decide who you meet. Publishing is free.